Deep learning-based brain transcriptomic signatures associated with the neuropathological and clinical severity of Alzheimer's disease

A review of neuroimaging-based data-driven approach for Alzheimer's disease heterogeneity analysis

Alzheimer's disease (AD) is a complex form of dementia and due to its high phenotypic variability, its diagnosis and monitoring can be quite challenging. Biomarkers play a crucial role in AD diagnosis and monitoring, but interpreting these biomarkers can be problematic due to their spatial and temporal heterogeneity. Therefore, researchers are increasingly turning to imaging-based biomarkers that employ data-driven computational approaches to examine the heterogeneity of AD. In this comprehensive review article, we aim to provide health professionals with a comprehensive view of past applications of data-driven computational approaches in studying AD heterogeneity and planning future research directions. We first define and offer basic insights into different categories of heterogeneity analysis, including spatial heterogeneity, temporal heterogeneity, and spatial-temporal heterogeneity. Then, we scrutinize 22 articles relating to spatial heterogeneity, 14 articles relating to temporal heterogeneity, and five articles relating to spatial-temporal heterogeneity, highlighting the strengths and limitations of these strategies. Furthermore, we discuss the importance of understanding spatial heterogeneity in AD subtypes and their clinical manifestations, biomarkers for abnormal orderings and AD stages, the recent advancements in spatial-temporal heterogeneity analysis for AD, and the emerging role of omics data integration in advancing personalized diagnosis and treatment for AD patients. By emphasizing the significance of understanding AD heterogeneity, we hope to stimulate further research in this field to facilitate the development of personalized interventions for AD patients.

scHybridBERT: integrating gene regulation and cell graph for spatiotemporal dynamics in single-cell clustering

Graph learning models have received increasing attention in the computational analysis of single-cell RNA sequencing (scRNA-seq) data. Compared with conventional deep neural networks, graph neural networks and language models have exhibited superior performance by extracting graph-structured data from raw gene count matrices. Established deep neural network-based clustering approaches generally focus on temporal expression patterns while ignoring inherent interactions at gene-level as well as cell-level, which could be regarded as spatial dynamics in single-cell data. Both gene-gene and cell-cell interactions are able to boost the performance of cell type detection, under the framework of multi-view modeling. In this study, spatiotemporal embedding and cell graphs are extracted to capture spatial dynamics at the molecular level. In order to enhance the accuracy of cell type detection, this study proposes the scHybridBERT architecture to conduct multi-view modeling of scRNA-seq data using extracted spatiotemporal patterns. In this scHybridBERT method, graph learning models are employed to deal with cell graphs and the Performer model employs spatiotemporal embeddings. Experimental outcomes about benchmark scRNA-seq datasets indicate that the proposed scHybridBERT method is able to enhance the accuracy of single-cell clustering tasks by integrating spatiotemporal embeddings and cell graphs.

Artificial intelligence for dementia genetics and omics

Genetics and omics studies of Alzheimer's disease and other dementia subtypes enhance our understanding of underlying mechanisms and pathways that can be targeted. We identified key remaining challenges: First, can we enhance genetic studies to address missing heritability? Can we identify reproducible omics signatures that differentiate between dementia subtypes? Can high-dimensional omics data identify improved biomarkers? How can genetics inform our understanding of causal status of dementia risk factors? And which biological processes are altered by dementia-related genetic variation? Artificial intelligence (AI) and machine learning approaches give us powerful new tools in helping us to tackle these challenges, and we review possible solutions and examples of best practice. However, their limitations also need to be considered, as well as the need for coordinated multidisciplinary research and diverse deeply phenotyped cohorts. Ultimately AI approaches improve our ability to interrogate genetics and omics data for precision dementia medicine. HIGHLIGHTS: We have identified five key challenges in dementia genetics and omics studies. AI can enable detection of undiscovered patterns in dementia genetics and omics data. Enhanced and more diverse genetics and omics datasets are still needed. Multidisciplinary collaborative efforts using AI can boost dementia research.

Genetically-regulated pathway-polygenic risk score (GRPa-PRS): A risk stratification method to identify genetically regulated pathways in polygenic diseases

Background

Alzheimer's disease (AD) is a common neurodegenerative disease in the elderly population, with genetic factors playing an important role. A considerable proportion of elderly people carry a high genetic AD risk but evade AD. On the other hand, some individuals with a low risk for AD eventually develop AD. We hypothesized that unknown counterfactors might be involved in reversing the polygenic risk scores (PRS) prediction, which might provide insights into AD pathogenesis, prevention, and early clinical intervention.

Methods

We built a novel computational framework to identify genetically-regulated pathways (GRPa) using PRS-based stratification for each cohort. We curated two AD cohorts with genotyping data; the discovery and the replication dataset include 2722 and 2492 individuals, respectively. First, we calculated the optimized PRS model based on the three latest AD GWAS summary statistics for each cohort. Then, we sub-grouped the individuals by their PRS and clinical diagnosis into groups such as cognitively normal (CN) with high PRS for AD (resilient group), AD cases with low PRS (susceptible group), and AD/CNs participants with similar PRS backgrounds. Lastly, we imputed the individual genetically-regulated expression (GReX) and identified the differential GRPas between subgroups with gene-set enrichment analysis and gene-set variational analysis in 2 models with and without the effect of APOE.

Results

For each subgroup, we conducted the same procedures in both the discovery and replication datasets across three PRS models for comparison. In Model 1 with the APOE region, we identified well-known AD-related pathways, including amyloid-beta clearance, tau protein binding, and astrocytes response to oxidative stress. In Model 2 without the APOE region, synapse function, microglia function, histidine metabolism, and thiolester hydrolase activity were significant, suggesting that they are pathways independent of the effect of APOE. Finally, our GRPa-PRS method reduces the false discovery rate in detecting differential pathways compared to another variants-based pathway PRS method.

Conclusions

We developed a framework, GRPa-PRS, to systematically explore the differential GRPas among individuals stratified by their estimated PRS. The GReX-level comparison among those groups unveiled new insights into the pathways associated with AD risk and resilience. Our framework can be extended to other polygenic complex diseases.

Unique Transcriptional Signatures Correlate with Behavioral and Psychological Symptom Domains in Alzheimer's Disease

Despite the significant burden, cost, and worse prognosis of Alzheimer's disease (AD) with behavioral and psychological symptoms of dementia (BPSD), little is known about the molecular causes of these symptoms. Using antemortem assessments of BPSD in AD, we demonstrate that individual BPSD can be grouped into 4 domain factors in our sample: affective, apathy, agitation, and psychosis. Then, we performed a transcriptome-wide analysis for each domain utilizing bulk RNA-seq of post-mortem anterior cingulate cortex (ACC) tissue. Though all 4 domains are associated with a predominantly downregulated pattern of hundreds of differentially expressed genes (DEGs), most DEGs are unique to each domain, with only 22 DEGs being common to all BPSD domains, including TIMP1. Weighted gene co-expression network analysis (WGCNA) yielded multiple transcriptional modules that were shared between BPSD domains or unique to each domain, and NetDecoder was used to analyze context-dependent information flow through the biological network. For the agitation domain, we found that all DEGs and a highly correlated transcriptional module were functionally enriched for ECM-related genes including TIMP1, TAGLN, and FLNA. Another unique transcriptional module also associated with the agitation domain was enriched with genes involved in post-synaptic signaling, including DRD1, PDE1B, CAMK4, and GABRA4. By comparing context-dependent changes in DEGs between cases and control networks, ESR1 and PARK2 were implicated as two high impact genes associated with agitation that mediated significant information flow through the biological network. Overall, our work establishes unique targets for future study of the biological mechanisms of BPSD and resultant drug development.

Deep Learning-Based Diagnosis of Alzheimer’s Disease

Alzheimer’s disease (AD), the most familiar type of dementia, is a severe concern in modern healthcare. Around 5.5 million people aged 65 and above have AD, and it is the sixth leading cause of mortality in the US. AD is an irreversible, degenerative brain disorder characterized by a loss of cognitive function and has no proven cure. Deep learning techniques have gained popularity in recent years, particularly in the domains of natural language processing and computer vision. Since 2014, these techniques have begun to achieve substantial consideration in AD diagnosis research, and the number of papers published in this arena is rising drastically. Deep learning techniques have been reported to be more accurate for AD diagnosis in comparison to conventional machine learning models. Motivated to explore the potential of deep learning in AD diagnosis, this study reviews the current state-of-the-art in AD diagnosis using deep learning. We summarize the most recent trends and findings using a thorough literature review. The study also explores the different biomarkers and datasets for AD diagnosis. Even though deep learning has shown promise in AD diagnosis, there are still several challenges that need to be addressed.

The Key Factors Predicting Dementia in Individuals With Alzheimer’s Disease-Type Pathology

Dementia affects millions of individuals worldwide, yet there are no effective treatments. Alzheimer’s disease, the most common form of dementia, is characterized by amyloid and tau pathology with amyloid accumulation thought to precipitate tau pathology, neurodegeneration, and dementia. The Religious Orders Study and Memory and Aging Project (ROSMAP) cohort is a unique resource with quantitative pathology from multiple brain regions, RNA sequencing, and longitudinal cognitive data. Our previous work applying machine learning to the RNA sequencing data identified lactoferrin (LTF) as the gene most predictive of amyloid accumulation with a potential amyloidogenic mechanism identified in vitro and with cell-culture models. In the present study, we examined which pathologies and genes were related to cognitive status (dementia, mild impairment, and no cognitive impairment) and rate of cognitive decline. Tau load in the anterior cingulate and ADAMTS2, encoding a metallopeptidase, were the respective regional pathology and gene most associated with cognitive decline, while PRTN3, encoding a serine protease, was the key protective feature. ADAMTS2, but not PRTN3, was related to amyloid and tau load in the previous study while LTF was not related to cognitive decline here. These findings confirm a general relationship between tau pathology and dementia, show the specific importance of tau pathology in the anterior cingulate cortex and identify ADAMTS2 as a potential target for slowing cognitive decline.